Random BSODs and other errors

June 11, 2009 at 00:54:51
Specs: Windows XP
Hey all, I really need some help; I've recently been getting random BSODs and memory-related errors when playing any game for roughly 30 minutes or more. Some of the many BSODs I have managed to achieve are:


0x0000008E (0xC0000005, 0xBD0C8429, 0xA7DE08E4, 0x00000000) nv4_disp.dll
0x0000000A (0xC0000010, 0x0000001C, 0x00000000, 0x804ffa41) win32k.sys
0x0000007F

The memory-related errors I've been getting are from any Source-based game (Half-Life 2, Team Fortress 2, Portal), and it says:


The instruction at "0x########" referenced memory at "0x########". The memory could no be "written".

Click on OK to terminate the program.

Where "0x########" randomly changes all the time. Most other games just hard crash or soft crash (such as The Sims 3 and Far Cry 2).

This is what I've done so far, to troubleshoot the error:

1. Got some new RAM (MemTest86 showed that my old memory was having trouble).
2. Swapped out a faulty hard-drive (bad sectors).
3. Changed to Windows 7 (from XP).
4. Removed one of my new RAM modules and tested them separately in different slots.
5. Increased the voltage for my RAM.
6. Yesterday I realised that I was running 800MHz RAM in a 667MHz motherboard. The motherboard specs said it could be overclocked to reach 800MHz safely, so I overclocked it with Fox Intelligent Overclock yesterday (it's been running the RAM at 667MHz since I got it).
7. I monitored system temperatures, which stayed low, even under load.

I did some other things as well, but I can't really remember. Anyway, my specs are:


Intel Core 2 Quad Q9300 @ 2.50GHz (4 CPUs)
OCZ Gold Vista 4GB (2x2GB) PC2-6400C5 800MHz Dual Channel Kit RAM
NVIDIA GeForce GTX 260
HIPER 730WATT, ATX V2.2
Motherboard: Foxconn M7VMX-K
Microsoft Windows XP Professional Service Pack 3

I haven't overclocked anything, except for my motherboard RAM frequency. I'm really not sure what else I can try, if anyone's got any ideas, I'd greatly appreciate it, thanks in advance.


See More: Random BSODs and other errors

Report •


#1
June 11, 2009 at 01:26:32
Can you post any of the mini-dumps from the crashes?

Anyhow, it sounds like a RAM issue. From your post, you seem to have had a lot of different issues with RAM, and I wouldn't be surprised if you still have some.

Are you absolutely sure your RAM is installed correctly and working correctly? Did you run MemTest on all of the new RAM modules, and allow it to run for a length of time.

-Ryan Adams
Free Computer Tips and more:http://RyanTAdams.com

Paid Tech Support: Black Diamond


Report •

#2
June 11, 2009 at 01:45:08
I'll have to re-enable dumping, it was turned off for some reason. I'll post the mini-dump for the next BSOD I get, which won't be long. Sorry.

I'm pretty sure it's installed correctly (any ideas on checking that though?), and I think it's working correctly. I'll start a MemTest on the new RAM in a second.

Thanks for the help!


Report •

#3
June 11, 2009 at 01:58:34
I just tried to start MemTest 3.5 and it gets to 48% of test #1 and restarts the PC. I then tried experimental version 3.5 (for multi CPUs) and it just freezes. Version 3.4 seems to be working, but it's a little weird that 3.5 didn't? Anyway, it's running at the moment, I'll let you know if I get any errors. Thanks again.

Report •

Related Solutions

#4
June 11, 2009 at 02:47:31
Sorry for the triple post, but the test has been running for nearly an hour now and test #7 has shown 539 errors. This can't be good. Is it guaranteed that my RAM is what's causing these errors or am I just unlucky enough to have purchased a faulty set?

I seem to remember that my previous RAM modules also showed errors on test #7 and all the other tests passed. Any ideas? Thanks again.


Report •

#5
June 11, 2009 at 05:11:24
The Q9300 is a 1333MHz FSB CPU. The correct speed to run the RAM is 667MHz DDR. Underclock your RAM & make sure the timings & voltage are set as per the manufacturer recommendation.

Report •

#6
June 11, 2009 at 08:27:36
I got the RAM from Novatech and they say the timings are 5-5-5-18, I had a look at what my BIOS was set to: 4-5-5-14 or something along those lines. I've now changed them to 5-5-5-18.

I also changed the voltage back up to 2.05V. Novatech says I need 2.1V and can go up to 2.2V without invalidating my warranty (2.1-2.2v VDIMM), but the OCZ website says I can only go up to 2.1V (and only 1.8V is needed), so I've kept it at 2.05V, because my BIOS only allows me to do 1.8V, 2.05V or 2.2V. Maybe I should change it to 2.2V and hope that I can still RMA them (if I need to) under what Novatech said.

I've also downclocked my DDR frequency to 667MHz (and the FSB is set to 1333MHz). After doing all that, I ran the test again and it still produced errors. Can 800MHz RAM run at 667MHz then? And should I just up my voltage to 2.2V and hope for the best?

Thanks again for the help.


Report •

#7
June 11, 2009 at 09:02:51
There is the possibility that it is actually your motherboard that is bad, and not the RAM, but my bet would still be on the RAM at this point.

Most motherboards will automatically configure the RAM, Processor, etc. unless you specifically override these settings with your own. What happens if you just let the motherboard decide what timing/speed/voltage to run the RAM at?

-Ryan Adams
Free Computer Tips and more:http://RyanTAdams.com

Paid Tech Support: Black Diamond


Report •

#8
June 11, 2009 at 10:06:50
When I set my BIOS to "optimal" memory timing settings it uses the following values: 4-5-5-14 and MemTest displays errors. The only thing I haven't tried is boosting the RAM to 2.2V (but that might break my warranty as it's not clear). Otherwise, it's beginning to look more and more like the motherboard. Anything else?

Report •

#9
June 12, 2009 at 02:41:14
I think I've fixed it. Last night I whipped the power supply out of another computer (because I was about to buy a new motherboard) and used it on mine: I only connected the bare minimum and then ran MemTest and I didn't receive any errors. I connected a few more things and this time MemTest did display errors.

I then disconnected everything and connected my original power supply. I connected each hard-drive to a different line of connectors and it received no errors. I've got 3 case fans, two 12cm and one 9cm, so I tried connecting these again and the errors returned. I removed one of my 12cm and ran MemTest for 20 hours without any errors.

I'm not quite sure why this works, but it looks like I was overloading the power supply. Anyway, thanks for your help RTAdams89 and jam, it was greatly appreciated.


Report •

#10
June 12, 2009 at 14:17:20
I seem to have sorted my RAM issues out, but now I've triggered a whole new line of errors: I just got another BSOD, but this time it said "IRQL_NOT_LESS_OR_EQUAL" and referenced the RtkHDAud.sys file. I suppose that's Realtek HD Audio, which is my on-board audio chip. Maybe my motherboard is playing up then? I uploaded the minidump though. Thanks again.

Report •

#11
June 14, 2009 at 13:31:46
I just received another blue screen, this time it was a 0x0000007F. Both of them occurred after playing games for a long period of time. Here's the minidump again.

Report •

#12
June 15, 2009 at 01:05:15
Well the first dump was related to an issue with the audio driver as you already knew. This latest one refers to memory corruption. You are definitely having issues with RAM, the trouble is figuring out if it is the RAM or motherboard. Can you find another PC to put the RAM in and test with MemTest? This will definatively implicate or exonerate the RAM itself.

-Ryan Adams
Free Computer Tips and more:http://RyanTAdams.com

Paid Tech Support: Black Diamond


Report •

#13
June 16, 2009 at 05:15:54
OK, I managed to test both RAM modules in another machine and MemTest returned no errors, so I can safely rule out the possibility of faulty RAM. I suppose that leads us to the motherboard; I might just go ahead and buy the Foxconn ELA. Any thoughts?

It's not likely to be anything to do with the graphics card though is it? I mean it only BSODs when I've been playing a game for a while. Just making sure I've ruled out everything else.

Thanks again!


Report •

#14
June 16, 2009 at 10:56:43
You might try resetting the motherboard BIOS to the defaults, disconnecting everything possible from the system (all USB devices, PCI/AGP add-in cards, drives, etc.) then run memtest again. If it still throws errors, the motherboard is bad. If it passes, add back devices one by one, retesting after each.

-Ryan Adams
Free Computer Tips and more:http://RyanTAdams.com

Paid Tech Support: Black Diamond


Report •

#15
June 21, 2009 at 09:11:43
OK, I did a CMOS reset with the on-board jumper and then ran a few tests, which showed no errors. I played a few games for hours without BSODs or crashes and so I thought I'd fixed it.

However, last night it started crashing in Team Fortress 2 again and then blue screened in Prototype (with 0x0000007F). So something is definitely wrong still. I'll try doing a CMOS reset and disconnecting everything later, but if it's stable for a while and then suddenly starts crashing again, it'll be hard to troubleshoot what could be causing it. I'll keep you updated, thanks again for the help.


Report •


Ask Question